There are powerful business tools that help you analyse the sequential flow of raw materials to storage facilities, then into manufacturing or fabrication, then into finished goods inventory, then to distribution, then to retail inventory, and finally to consumers. Each holding point is an inventory that can be analysed individually, optimising things such as shortages, surpluses, and holding and acquisition costs; and this is inventory optimisation, writes John W. Boudreau in ‘Retooling HR: Using proven business tools to make better decisions about talent’ (Harvard).
HR (human resource) and its stakeholders constantly confront analogous workforce inventory questions, he adds. Examples of such questions are: ‘Is employee turnover always bad? Should open jobs be filled as quickly as possible? Should some roles have talent surpluses, while others tolerate occasional talent shortages?’
Turnover prediction
Employee separations are called ‘turnover’ because they are like the situation where inventories turn over when they are depleted (after being sold or lost or spoiled) and then are replaced, notes Boudreau. The tools that have proved useful for managing other inventories reveal untapped potential for improved collaboration and accountability when it comes to vital questions about employee turnover, surpluses, and shortages, he assures.
Interestingly, one learns about Google’s formula that predicts the probability that each employee will leave, as if to help the company ‘get inside people’s heads even before they know they might leave’! “Ameriprise Financial has developed similar formulas and reports the probability for each individual who works for certain unit leaders, in hopes of helping the managers target their retention efforts.”
Among the cases cited in the book is one about how Microsoft’s Bill Gates and Steve Ballmer tolerated employee shortages, insisting on employing fewer employees than required – as per the formula ‘n minus 1’ where n was the number actually needed. And their argument was that the shortage would motivate the leaders not to settle for anyone but the best, and convey the idea that the company’s employees were expected to excel.
On-demand workforce
An example discussed elaborately is that of IBM Global Services, in the context of workforce inventory and the on-demand workforce. The problem statement is simple: “To win contracts, IBM must guarantee that it can assemble a team having the necessary skills on time and on budget. It must balance people who may work on several projects with those working on one, and people who are needed long term versus those needed short term. IBM doesn’t know until the contract is signed precisely what the needed workforce will do.”
The solution is in the form of a system that can deliver talent on demand as contract needs arise, informs the author. He explains how, to enable this agile supply of talent, IBM leaders draw from workforce inventories that include not only employees but also contractors.
Expertise taxonomy
“To provide a common language allowing every business in the world to describe its talent supply and demand using the same language, IBM has developed an expertise taxonomy, essentially making every country and job a searchable talent inventory. ‘Talent on demand’ is a term IBM uses to capture the idea of reframing the question from filling vacancies to optimising talent availability.”
Since the costs of talent shortages could be very high, perhaps even the loss of million-dollar contracts, IBM invested heavily in accurate predictions – and in a global workforce taxonomy that cost perhaps $100 million – to make its internal talent inventories as transparent and accessible as possible, reasons Boudreau.
Conjoint analysis
A chapter on ‘segmentation research for talent’ delves into ‘conjoint analysis to optimise total rewards’ at Microsoft.
Conjoint analysis, for starters, ‘sees products and services as combinations of attributes.’ For example, credit cards have attributes such as interest rates, annual fees, credit limits, the ability to put your picture on the card, and the reputation of the issuing company, the author instructs. Similarly, ‘personal computers have attributes such as monitor size and refresh rate, keyboard look and feel, operating system vulnerability to viruses, and the reputation and hipness of the computer brand.’
As the dot-com boom was waning, companies like Microsoft were rethinking their traditional reliance on stock options and other equity-based rewards, Boudreau narrates. He quotes a snatch from Ballmer’s interview for ‘Fast Company,’ thus: “We’re spending much more time focusing on the quality of the job. We’re thinking hard about how to keep jobs big and full of impact. That’s the key: doing more than just fixating on compensation.”
Comprehensive total rewards strategy
The Redmond giant, therefore, had its HR and total rewards leaders work with Towers Perrin to apply conjoint analysis, with the goal of exploring ‘a comprehensive total rewards strategy that not only would better accommodate the preferences of Microsoft’s employees but also would be administratively practical by helping Microsoft optimise where it customised and where it standardised.’
Unlike as in product-based conjoint analysis, the study here went beyond physical attributes to capture a wide variety of work experiences, including improved managerial effectiveness, the author recounts.
Conceding that it might be harder to provide precise costs for something like managerial effectiveness, and that employees might interpret the phrase somewhat differently, he adds that these imperfections were far less important than the need to know whether the employees valued improved managerial effectiveness highly compared to more tangible rewards.
Adaptive questioning
“Microsoft took advantage of computerised tools that allow adaptive questioning: the computer presents users with attributes to determine what is important to each individual. Then it constructs questions tailored to each individual’s answers and continues to adapt as the person answers further questions.”
Importantly, the final part of the survey asked employees to estimate the likelihood they would stay with the company for a specified period if they received the rewards in their preferred package.
Projected return on investment
Turnover intentions are not a perfect predictor of actual turnover, but they are correlated with actual turnover, the author observes. He finds also that asking about turnover intentions offered Microsoft a significant additional opportunity: the chance to connect investments in rewards with the costs of turnover and thus optimise total rewards based on projected return on investment.
“In collaboration with their finance colleagues, the Microsoft team estimated the cost of turnover as well as the cost of each of the reward levels… Calculating these costs involved applying standard financial and behavioural costing methods.”
The payoff, as Boudreau concludes, was that Microsoft could calculate the monetary effect of its decisions and identify optimum alternatives to its current approach. “Borrowing principles from financial analysis, the team chose to be conservative in estimating turnover cost savings and liberal in estimating reward programme cost expenditures…”
Recommended addition to professional HR managers’ shelf.
**
Tailpiece
“We fired our HR chief when we found…”
“That he was causing human resource bottlenecks?”
“No, our attrition tracking system popped up his name as the key cause!”
**
BookPeek.blogspot.com
